MULTISCALE LOGGING WHILE DRILLING DATA INTEGRATION UNVEILED COMPLEX GEOLOGICAL SCENARIO WHILE GEOSTEERING

The first Eni geosteering operation in Mexico was executed during the global COVID-19 crisis. The complex geology and the uncertainty related to this undrilled portion of the reservoir determined the employment of advanced Logging While Drilling (LWD) technology for real-time geosteering and a comprehensive geological interpretation. The target is an oil bearing sandstone reservoir, represented by deltaic front sands bars within an anticline structure on a salt core with faults and lateral heterogeneity. A sedimentological conceptual model was used to feed the 3D geological model, supporting a development strategy based on the geosteering of a horizontal well. The trajectory was designed within the best petrophysical properties interval to maximize production. The pre-drill risk analysis determined the need for a pilot hole to confirm structural setting, reservoir properties and fluid contacts to mitigate the associated uncertainties. The landing data acquisition strategy included standard LWD measurements and density images to optimize the wellbore inclination. The drain section was going to be geosteered with an Ultra-Deep Azimuthal Electromagnetic tool, dual-physics imager for oil-based mud systems and sourceless Density/Neutron technology. The pilot hole confirmed the pre-drill expected scenario but the LWD images and data interpreted while landing, revealed a more complex than expected target reservoir architecture. The detailed geological picture was completed while drilling the drain section. The multi-scale data (Reservoir Mapping information, Resistivity images, Logs, Seismic Interpretation and Pressure points) were integrated and exchanged 24/7 by experts through a commercial hub for team collaboration. A communication and information sharing protocol was customized to overcome the restrictions dictated by COVID-19 health emergency. The combination of acquired information and knowledge, unveiled a reservoir made of stacked clinoforms with internal geometries non-conformable with the general structural trend. Real-time geosteering with advanced technologies information, mitigated the impact of the unexpected complex subsurface setting. A total of 270 m were drilled inside the target, maximizing the drilled Net-to-Gross compared with the planned trajectory. Furthermore, the geological scenario reconstructed with multiscale LWD data, was exploited for a detailed 3D reservoir model update.

Managing Uncertainty in Geological Scenarios Using Machine Learning-Based Classification Model on Production Data

Training image (TI) has a great influence on reservoir modeling as a spatial correlation in the multipoint geostatistics. Unlike the variogram of the two-point geostatistics that is mathematically defined, there is a high degree of geological uncertainty to determine a proper TI. The goal of this study is to develop a classification model for determining the proper geological scenario among plausible TIs by using machine learning methods: (a) support vector machine (SVM), (b) artificial neural network (ANN), and (c) convolutional neural network (CNN). After simulated production data are used to train the classification model, the most possible TI can be selected when the observed production responses are put into the trained model. This study, as far as we know, is the first application of CNN in which production history data are composed as a matrix form for use as an input image. The training data are set to cover various production trends to make the machine learning models more reliable. Therefore, a total of 800 channelized reservoirs were generated from four TIs, which have different channel directions to consider geological uncertainty. We divided them into training, validation, and test sets of 576, 144, and 80, respectively. The input layer comprised 800 production data, i.e., oil production rates and water cuts for eight production wells over 50 time steps, and the output layer consisted of a probability vector for each TI. The SVM and CNN models reasonably reduced the uncertainty in modeling the facies distribution based on the reliable probability for each TI. Even though the ANN and CNN had roughly the same number of parameters, the CNN outperformed the ANN in terms of both validation and test sets. The CNN successfully classified the reference model’s TI with about 95% probability. This is because the CNN can grasp the overall trend of production history. The probabilities of TI from the SVM and CNN were applied to regenerate more reliable reservoir models using the concept of TI rejection and reduced the uncertainty in the geological scenario successfully.